• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.149-158
• /
• 2004

This study proposes a method for segmentation and recognition of traffic signs using shape information and edge image in real image. It first segments traffic sign candidate regions by connected component algorithm from binary images, obtained by utilizing the RGB color ratio of each pixel in the image, and then extracts actual traffic signs based on their symmetries on X- and Y-axes. Histogram equalization is performed for unsegmented candidate regions caused by low contrast in the image. In the recognition stage, it utilizes shape information including projection profiles on X- and Y-axes, moment, and the number of crossings and distance which concentric circular patterns and 8-directional rays from region center intersects with edges of traffic signs. It finally performs recognition by measuring similarity with the templates in the database. It will be shown from several experimental results that the system is robust to environmental factors, such as light and weather condition.

• Smart Media Journal
• /
• v.5 no.1
• /
• pp.38-43
• /
• 2016

As smart environment is spread out in our living environments, the needs of an approach related to Human Computer Interaction(HCI) is increases. One of them is head pose estimation. it related to gaze direction estimation, since head has a close relationship to eyes by the body structure. It's a key factor in identifying person's intention or the target of interest, hence it is an essential research in HCI. In this paper, we propose an approach for head pose estimation with pre-defined several directions by random forest classifier. We use canny edge detector to extract feature of the different facial image which is obtained between input image and averaged frontal facial image for extraction of rotation information of input image. From that, we obtain the binary edge image, and make two accumulated histograms which are obtained by counting the number of pixel which has non-zero value along each of the axes. This two accumulated histograms are used to feature of the facial image. We use CAS-PEAL-R1 Dataset for training and testing to random forest classifier, and obtained 80.6% accuracy.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.4
• /
• pp.239-244
• /
• 2009

Image contrast enhancement has an important role in image processing applications. Conventional contrast enhancement techniques, histogram stretching and histogram equalization, and many methods based on histogram equalization often fail to produce satisfactory results for broad variety of low-contrast images. So, this paper proposes a new image contrast enhancement method based on the clustering method. The number of cluster of histogram is found by analysing the histogram of original image. The histogram components is classified using K-means algorithm. And then these histogram components are performed histogram stretching and histogram equalization selectively by comparing cluster range with pixel rate of cluster. From the expremental results, the proposed method was more effective than conventional contrast enhancement techniques.

• Journal of KIISE:Software and Applications
• /
• v.27 no.1
• /
• pp.74-83
• /
• 2000

There is a rapid increase in the use of digital video information in recent years. Especially, user requires the environment which retrieves video from passive access to active access, to be more efficiently. we need to implement video retrieval system including video parsing, clustering, and browsing to satisfy user's requirement. In this paper, we first divide video sequence to shots which are primary unit for automatic indexing, using a hybrid method with mixing histogram method and pixel-based method. After the shot boundaries are detected, corresponding key frames can be extracted. Key frames are very important portion because they help to understand overall contents of video. In this paper, we first analyze camera operation in video and then select different number of key frames depend on shot complexity. At last, we compose panorama images from shots which are containing panning or tilting in order to provide more useful and understandable browsing environment to users.

• Korean Journal of Remote Sensing
• /
• v.16 no.1
• /
• pp.73-86
• /
• 2000

Commonly-used methods for camera modelling of pushbroom images were implemented and their performances were assessed. The models include Vector Propagation) model, Gugan and Downman(GD)'s model, Orun and Natarajan(ON)'s model, and Direct Linear Transformation(DLT) model The models were tested on a SPOT full-scene over Seoul. The number of ground control points(GCP) used range from 1 to 23. For less than 6 GCPs all other models fail except VP, with VP's accuracy being 2.7 pixels. With mode than 6 GCPs ON shows the best accuracy with 1pixel accuracy while the accuracy of VP is 1.5 pixels. GD fails in most cases due to the correlation among model parameters. The accuracy of DLT does not converge but fluctuates between 1 and 4 pixels subject to GCPs used. VP has an advantage in that its results can be used for the estimation of satellite orbit. Unresolved topics are: to remove errors in GCPs from the aforementioned accuracy value; to improve the performance of VP.

• Agribusiness and Information Management
• /
• v.10 no.2
• /
• pp.1-7
• /
• 2018

In the present day context of changing information needs of the farmers and diversified production systems there is an urgent need to look for the effective extension support system for the small and marginal farmers in the developing countries like India. The rapid developments in the collection and analysis of field data by using the spatial technologies like GPS&GIS were made available for the extension functionaries and clientele for the diversified information needs. This article describes the GIS and GPS based decision support system in precision agriculture for the resource poor farmers. Precision farming techniques are employed to increase yield, reduce production costs, and minimize negative impacts to the environment. The parameters those can affect the crop yields, anomalous factors and variations in management practices can be evaluated through this GPS and GIS based applications. The spatial visualisation capabilities of GIS technology interfaced with a relational database provide an effective method for analysing and displaying the impacts of Extension education and outreach projects for small and marginal farmers in precision agriculture. This approach mainly benefits from the emergence and convergence of several technologies, including the Global Positioning System (GPS), geographic information system (GIS), miniaturised computer components, automatic control, in-field and remote sensing, mobile computing, advanced information processing, and telecommunications. The PPP convergence of person (farmer), project (the operational field) and pixel (the digital images related to the field and the crop grown in the field) will better be addressed by this decision support model. So the convergence and emergence of such information will further pave the way for categorisation and grouping of the production systems for the better extension delivery. In a big country like India where the farmers and holdings are many in number and diversified categorically such grouping is inevitable and also economical. With this premise an attempt has been made to develop a precision farming model suitable for the developing countries like India.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.5
• /
• pp.2698-2717
• /
• 2019

This paper proposes a multi-stage encryption technique to enhance the level of secrecy of image to facilitate its secured transmission through the public network. A great number of researches have been done on image secrecy. The existing image encryption techniques like visual cryptography (VC), steganography, watermarking etc. while are applied individually, usually they cannot provide unbreakable secrecy. In this paper, through combining several separate techniques, a hybrid multi-stage encryption technique is proposed which provides nearly unbreakable image secrecy, while the encryption/decryption time remains almost the same of the exiting techniques. The technique consecutively exploits VC, steganography and one time pad (OTP). At first it encrypts the input image using VC, i.e., splits the pixels of the input image into multiple shares to make it unpredictable. Then after the pixel to binary conversion within each share, the exploitation of steganography detects the least significant bits (LSBs) from each chunk within each share. At last, OTP encryption technique is applied on LSBs along with randomly generated OTP secret key to generate the ultimate cipher image. Besides, prior to sending the OTP key to the receiver, first it is converted from binary to integer and then an asymmetric cryptosystem is applied to encrypt it and thereby the key is delivered securely. Finally, the outcome, the time requirement of encryption and decryption, the security and statistical analyses of the proposed technique are evaluated and compared with existing techniques.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.3917-3941
• /
• 2019

The recent interest in artificial intelligence and machine learning has partly contributed to an interest in the use of such approaches for hyperspectral remote sensing (HRS) imagery classification, as evidenced by the increasing number of deep framework with deep convolutional neural networks (CNN) structures proposed in the literature. In these approaches, the assumption of obtaining high quality deep features by using CNN is not always easy and efficient because of the complex data distribution and the limited sample size. In this paper, conventional handcrafted learning-based multi features based on expert knowledge are introduced as the input of a special designed CNN to improve the pixel description and classification performance of HRS imagery. The introduction of these handcrafted features can reduce the complexity of the original HRS data and reduce the sample requirements by eliminating redundant information and improving the starting point of deep feature training. It also provides some concise and effective features that are not readily available from direct training with CNN. Evaluations using three public HRS datasets demonstrate the utility of our proposed method in HRS classification.

• Korean Journal of Optics and Photonics
• /
• v.33 no.6
• /
• pp.303-309
• /
• 2022

Since the pixel size of the image sensor used in optical systems is gradually decreasing, the resolution specification of the optical system should be inevitably higher. If aberration change according to the eccentricity of a specific lens group occurs, only the amount of eccentricity of a specific lens group may be calculated with the traditional resolution adjustment method so that the aberration of the optical system is minimized to a certain extent. As a result, it is possible to increase the resolution of the optical system and to respond to a sensor with a large number of pixels. However, in the traditional method, there should be no change in specific aberration due to the eccentricity of a specific lens group. In this paper, we propose a new method to eliminate such a limitation of the traditional method in a camera optical system with a floating system, which is to choose and control the arbitrary two lens groups to easily minimize the eccentricity of the optical system in order to obtain an optical system with high resolution.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1520-1527
• /
• 2019

Superpixel methods are widely used in the preprocessing stage as a method to reduce computational complexity by simplifying images while maintaining the characteristics of images in the field of computer vision. It is common to generate superpixels with a regular size and form based on the pixel values rather than considering the characteristics of the image. In this paper, we propose a method to generate superpixels considering the characteristics of an image according to the application. The proposed method consists of two steps, and the first step is to oversegment an image so that the boundary information of the image is well preserved. In the second step, superpixels are merged based on similarity to produce the desired number of superpixels, where the form of superpixels are controlled by limiting the maximum size of superpixels. Experimental results show that the proposed method preserves the boundaries of an image more accurately than the existing method.